Venkatesh Rao @vgr Tweets are 90% vacuous views, apathetically held. Mediocritopian. May. 05, 2019 7 min read

A megatrend hypothesis inspired by several microtrend that I think are related:

a) waldenponding

b) rise of heavy duty information management methods like @fortelabs BASB (build a second brain)

c) conversational media eating authorial media

d) "hivebrain" jokes/references

If you put it together, you reach a megatrend conclusion: if you do not build a second brain or go offline, you will BECOME the second brain

What do I mean. Basically, there's no way to actually handle the volume of information and news that all of us appear to be handling right now. Which means we are getting augmented cognition resources from somewhere. The default place is "social" media.

The ersatz "social" functions of social media are not actually meeting social needs. Yet we indulge in them heavily.

Hypothesis 1: Evil addictive UX dark pattern grumble grumble ads hacked muh brain I'm going living in a log cabin

Hypothesis 2: They are meeting other needs

There's been a round of pretentious commentary in old media about how Twitter isn't actually very representative blah blah blah. They're missing the point. Those who endure and stay on here are neither meeting social needs, nor imagining they are "representative" of zeitgeist

What those of us who are here are doing is making a deal with the devil (or an angel): in return for being 1-2 years ahead of curve, we play 2nd brain to a shared first brain. We've ceded control of executive attention not to evil companies, but... an emergent oracular brain

I called it playing your part in the Global Social Computer in the Cloud (GSCITC). This has now become a much more important part of the argument I'm developing. If you don't build your own 2nd brain, the GSCITC *is* your first brain and the one in your skull is your 2nd brain.

What can we say about the GSCITC?

1. It is neither good nor evil, it just is

2. It is not smarter or dumber than humans (ie not an AGI or anything), just different

3. Its primary superpower is knowing the future

4. It is comparable to things like "the market"

Ccentral trade-off in managing your participation in GSCITC is: The more you attempt to consciously curate your participation rather than letting it set your priorities, the less oracular power you get in return. Waldenponding, BASB, and unmanaged use are 3 points on a spectrum.

If you choose to unplug completely (and or, do a draconian discipline of extremely tightly curated following, time-budgeting etc) you get zero temporal leverage.

At best you'll live int he same present as say readers of the NYT or viewers of Fox News.

If you build a high-maintenance 2nd-brain (whether it is making Anki cards out of everything, or using Tiago's BASB model), you'll give up some temporal leverage in return for 1st brain agency. You get (say) 1 year future-vision but less personal agency than waldenponders.

And finally, if you choose to be completely unmanaged in your use of social media, letting the winds of The Discourse drive what you do, you give up first-brain agency, but in return you get as much oracular power as the GSCITC as it has available to deliver. The Gonzo option.

You can actually extrapolate this model backwards through old media. If waldenponding puts you at zero temporal leverage, relying purely on (say) The Economist, which is about 6 months lag, goes into negative territory.

In general, temporal leverage can range from +2 years to -3000 years (if all you read are ancient Greek classics). The more reactionary you get, the more you need some *other* source of leverage to make up agency deficit. That wallowing in Greek classics better generate alpha.

Can you barbell your way out of this tradeoff? I like barbell thinking for many things but not for temporal agency. A portfolio of 90% greek classics and 10% shitposts just isn't going to do the trick... because the oracular agency of GSCITC participation has sharp thresholds

This addresses my main problem with waldenponding. To the extent *any* kind of thinking requires an input stream to work with, you can't arbitrarily decide a certain degree of retreat from "live" and a certain "deep work project" will be a net positive returns portfolio.

I have never had a problem with social media being a "distraction". My gonzo retreat/approach radar for intuitively calibrating how to balance my temporal leverage portfolio has been pretty good. And more importantly, requires no "addiction management" type behaviors.

I find that when I do get sucked into a longer project, I naturally retreat exactly as much as needed to "feed" that project the info stream it needs. If I need to spend more time with 10-year-old books I naturally do that.

This isn't a superpower. It's evidence the addiction hypothesis is much weaker than it looks. And that for nearly any subject, there is a shit ton more signal in the GSCITC than you might assume. Sign: output from "deep work" is generally very underwhelming.

I have been generally very unimpressed with the work people seem to generate when they go waldenponding to work on supposedly important things. The comparable people who stay more plugged in seem to produce better work.

My kindest reading of people who retreat so far it actually compromises their work is that it is a mental health preservation move because they can't handle the optimum GSCITC immersion for their project. Their work could be improved if they had the stomach for more gonzo-nausea.

My harshest reading is that they're narcissistic snowflakes who overvalue their work simply because they did it.

So my recommendations would be: develop an intuitive ability to tune your GSCITC immersion from full-retreat if you got a big grant to do some work only 3 other people in the world understand... to full-speed-ahead gonzo immersion if you need 2y oracular foresight

If you find your natural brain has a comfort zone, then build a second brain that equilibrates it there, and use it more or less as you need to wander from that set-point.

One reason I haven't invested in building as strong a BASB as many is simply that my comfort set-point is mostly near the gonzo edge. I need the full 2+ year foresight potential of the GSCITC for the thinking I'm most effective at. If it means continuous partial PTSD so be it.

One sign of the cost of this tradeoff. Though I can see this thread would be nice to put into a blog post, I don't feel like I can conjure up the energy to do it. It will likely remain in this context, only accessible and grokkable to people who are also in a roughly gonzo place

And also, I need a better term than GSCITC... but something without baggage. Hive mind, borg, AGI, efficient narrative marketplace... all the terms I've been able to think of have inappropriate or misleading connotations.

For those who came in late... this train of thought began with my Against Waldenponding newsletter last year  https://mailchi.mp/ribbonfarm/against-waldenponding 

Which I recently added a Part 2 to...  https://us1.campaign-archive.com/?u=78cbbb7f2882629a5157fa593&id=098f100576 

The Waldenponding newsletters were Oct 2018 and April 2019. I think I first broached this topic even earlier in this newsletter. Your Passport to the Metabrain (Jan 2017). Shit, this has been on my 2nd brain for a long time.  https://us1.campaign-archive.com/?u=78cbbb7f2882629a5157fa593&id=0f299f9100 

The cost of being in gonzo mode online is admittedly quite high. The effect on more retreated modes of production (essays, books, products) is very noticeable. My best thinking in the last few years has been on twitter, but the reach is far lower than my blog or newsletter

Final point: you are almost certainly overvaluing information from history and undervaluing information being processed live in the GSCITC for *any* project you take on. Why do I think that? At some point AlphaGoZero discarded all human history of Go playing and went full gonzo

This is a form of temporal egotism. We have identities anchored in personal/collective history, and we like to think distribution we are attached to is the distribution of input value. Chances are, an AI approaching any given problem you take on will throw away most of history.

Final final point. I flagged conversational media eating authorial media. This has been my personal ego challenge. It is hard to admit, having spent 10y building up an identity/vanity around my longform, that my conversational thinking embedded in the GSCITC is better quality.


You can follow @vgr.



Bookmark

____
Tip: mention @threader_app on a Twitter thread with the keyword “compile” to get a link to it.

Enjoy Threader? Sign up.

Threader is an independent project created by only two developers. The site gets 500,000+ visits a month and our iOS Twitter client was featured as an App of the Day by Apple. Running this space is expensive and time consuming. If you find Threader useful, please consider supporting us to make it a sustainable project.